Análisis 5: pEhExvsUmasM
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
load('CheackPointOne.RData')Quinta Clasificacion: pEhExvsUmasM
head(pEhExvsUmasM,10);## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199
## 7 EHI_000290A 26.39785 19.65360 70.24882 18.54395 14.10307 81.15238
## 8 EHI_000300A 80.39344 115.21078 15.61085 133.51645 115.17504 31.59916
## 9 EHI_000410A 20.39834 24.39758 138.93656 28.18681 24.68037 65.35280
## 10 EHI_000430A 31.19746 35.91866 14.04976 25.96153 14.10307 10.77244nbreaks <- 10
data5 <- pEhExvsUmasM; head(data5)## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199Log-Normalización
sample1 <- data5$pEhEx_1; sample2 <- data5$pEhEx_2; sample3 <- data5$pEhEx_3;
samplevs1 <- data5$UmasM_1; samplevs2 <- data5$UmasM_2; samplevs3 <- data5$UmasM_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsumasM1 <- log2(samplevs1+1)
log2samplevsumasM2 <- log2(samplevs2+1); log2samplevsumasM3 <- log2(samplevs3+1)
data5 <- cbind(data5, log2sample1,log2sample2,log2sample3,
log2samplevsumasM1,log2samplevsumasM2,log2samplevsumasM3)
head(data5)## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199
## log2sample1 log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## 1 7.459882 8.759868 8.904774 5.749419 8.745886
## 2 8.467919 8.417110 5.314000 6.889844 8.538024
## 3 10.199605 10.916716 11.052005 9.540979 10.187333
## 4 8.348241 8.356324 9.365349 9.462180 7.740125
## 5 6.781864 6.087241 5.680082 6.904367 6.603402
## 6 5.804521 6.399908 5.971819 5.872713 5.922163
## log2samplevsumasM3
## 1 9.778619
## 2 5.482538
## 3 12.794720
## 4 10.625030
## 5 7.128629
## 6 5.143388save.image('CheckPointFifth.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')library(ggplot2);library(dplyr);library("fitdistrplus");##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union## Loading required package: MASS##
## Attaching package: 'MASS'## The following object is masked from 'package:dplyr':
##
## select## Loading required package: survivallibrary("MASS");library("survival")head(data5)## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199
## log2sample1 log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## 1 7.459882 8.759868 8.904774 5.749419 8.745886
## 2 8.467919 8.417110 5.314000 6.889844 8.538024
## 3 10.199605 10.916716 11.052005 9.540979 10.187333
## 4 8.348241 8.356324 9.365349 9.462180 7.740125
## 5 6.781864 6.087241 5.680082 6.904367 6.603402
## 6 5.804521 6.399908 5.971819 5.872713 5.922163
## log2samplevsumasM3
## 1 9.778619
## 2 5.482538
## 3 12.794720
## 4 10.625030
## 5 7.128629
## 6 5.143388Muestra 1
log2sample1 <- data5$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))## [1] 6.24652## [1] 2.884429head(log2sample1,5)## [1] 7.459882 8.467919 10.199605 8.348241 6.781864summary(data5)## GenId UmasM_1 UmasM_2 UmasM_3
## Length:4919 Min. : 0.0 Min. : 0.00 Min. : 0.0
## Class :character 1st Qu.: 16.8 1st Qu.: 18.30 1st Qu.: 14.0
## Mode :character Median : 46.8 Median : 52.18 Median : 54.6
## Mean : 1915.1 Mean : 1003.67 Mean : 3444.8
## 3rd Qu.: 198.0 3rd Qu.: 193.15 3rd Qu.: 295.0
## Max. :340994.2 Max. :145896.83 Max. :1488475.8
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.00 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 16.32 1st Qu.: 15.28 1st Qu.: 15.1 1st Qu.: 4.114
## Median : 48.96 Median : 48.19 Median : 52.4 Median : 5.643
## Mean : 1394.44 Mean : 1752.18 Mean : 1898.2 Mean : 6.247
## 3rd Qu.: 199.53 3rd Qu.: 222.71 3rd Qu.: 231.2 3rd Qu.: 7.648
## Max. :213518.02 Max. :279409.95 Max. :724577.3 Max. :17.704
## log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 4.025 1st Qu.: 4.007 1st Qu.: 4.154 1st Qu.: 4.270
## Median : 5.620 Median : 5.739 Median : 5.579 Median : 5.733
## Mean : 6.187 Mean : 6.152 Mean : 6.222 Mean : 6.256
## 3rd Qu.: 7.805 3rd Qu.: 7.860 3rd Qu.: 7.637 3rd Qu.: 7.601
## Max. :18.092 Max. :19.467 Max. :18.379 Max. :17.155
## log2samplevsumasM3
## Min. : 0.000
## 1st Qu.: 3.912
## Median : 5.798
## Mean : 6.317
## 3rd Qu.: 8.210
## Max. :20.505ndata5 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)## [1] 6.24652StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)## [1] 2.884429Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)## [1] 0.4206592 0.7701346 1.3704911 0.7286438 0.1855978 -0.1532363tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.310539e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8877976 0.8234548
## 70 -0.9174931 0.9668385
## 75 -0.9827617 1.1346553
## 80 -1.0188953 1.3615138
## 85 -1.1000839 1.6923518
## 90 -1.1462228 2.1593960
## 95 -1.3906263 2.6069712
## 99 -2.1655999 3.1547856col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
Muestra 2
log2sample2 <- data5$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))## [1] 6.18657## [1] 3.14197head(log2sample2,5)## [1] 8.759868 8.417110 10.916716 8.356324 6.087241ndata5 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)## [1] 6.18657StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)## [1] 3.14197Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)## [1] 0.81900802 0.70991786 1.50547137 0.69057145 -0.03161338 0.06789964tst<- Normlog2sample2hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.093861e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995; CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8936351 0.8467827
## 70 -0.8936351 0.9884290
## 75 -1.0112465 1.1344069
## 80 -1.0836965 1.3626690
## 85 -1.1697595 1.6777371
## 90 -1.2757625 2.1128235
## 95 -1.6121684 2.5115181
## 99 -1.9690098 3.0656774col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
Muestra 3
log2sample3 <- data5$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))## [1] 6.152478## [1] 3.158081head(log2sample3,5)## [1] 8.904774 5.314000 11.052005 9.365349 5.680082ndata5 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)## [1] 6.152478StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)## [1] 3.158081Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)## [1] 0.87150918 -0.26550237 1.55142547 1.01734948 -0.14958317 -0.05720518tst<- Normlog2sample3hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.292160e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1.00000e+00 -3.26782e-11
## sd -3.26782e-11 1.00000e+00par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8811857 0.8359346
## 70 -0.9495345 0.9646852
## 75 -1.0299396 1.1073525
## 80 -1.1855441 1.3225658
## 85 -1.2519401 1.6081997
## 90 -1.4233556 2.0263238
## 95 -1.7009100 2.4893700
## 99 -1.9481699 3.1039165Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
Muestra pEhExvsumasM1
log2vsumasM1 <- data5$log2samplevsumasM1; head(mean(log2vsumasM1)); head(sd(log2vsumasM1))## [1] 6.221901## [1] 2.955459head(log2vsumasM1,5)## [1] 5.749419 6.889844 9.540979 9.462180 6.904367ndata5 <- length(log2vsumasM1)
hist(log2vsumasM1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM1')
meanlog2vsumasM1 <- mean(log2vsumasM1); head(meanlog2vsumasM1)## [1] 6.221901StdDevlog2vsumasM1 <- sd(log2vsumasM1); head(StdDevlog2vsumasM1)## [1] 2.955459Normlog2vsumasM1 <- (log2vsumasM1-meanlog2vsumasM1)/StdDevlog2vsumasM1; head(Normlog2vsumasM1)## [1] -0.1598676 0.2260030 1.1230332 1.0963710 0.2309173 -0.1181501tst<- Normlog2vsumasM1Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -9.853035e-17 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8531906 0.8236011
## 70 -0.9004672 0.9332295
## 75 -0.9528182 1.1118993
## 80 -0.9528182 1.3280269
## 85 -1.0781339 1.6545578
## 90 -1.1553694 2.1045904
## 95 -1.3603164 2.6974724
## 99 -2.1052233 3.3025359Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
Muestra pEhExvsumasM2
log2vsumasM2 <- data5$log2samplevsumasM2; head(mean(log2vsumasM2)); head(sd(log2vsumasM2))## [1] 6.255602## [1] 2.72471head(log2vsumasM2,5)## [1] 8.745886 8.538024 10.187333 7.740125 6.603402Primer Histograma
ndata5 <- length(log2vsumasM2)
hist(log2vsumasM2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM2')
meanlog2vsumasM2 <- mean(log2vsumasM2); head(meanlog2vsumasM2)## [1] 6.255602StdDevlog2vsumasM2 <- sd(log2vsumasM2); head(StdDevlog2vsumasM2)## [1] 2.72471Normlog2vsumasM2 <- (log2vsumasM2-meanlog2vsumasM2)/StdDevlog2vsumasM2; head(Normlog2vsumasM2)## [1] 0.9139630 0.8376754 1.4429905 0.5448372 0.1276466 -0.1223761tst<- Normlog2vsumasM2** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM2',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.626526e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8779788 0.8200071
## 70 -0.9297388 0.9699103
## 75 -0.9871123 1.1247423
## 80 -1.0514667 1.3424300
## 85 -1.1247392 1.7188713
## 90 -1.2098054 2.1841922
## 95 -1.3112040 2.5940822
## 99 -1.8422575 3.1504680Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
Muestra pEhExvsumasM3
log2vsumasM3 <- data5$log2samplevsumasM3; head(mean(log2vsumasM3)); head(sd(log2vsumasM3))## [1] 6.317092## [1] 3.410882head(log2vsumasM3,5)## [1] 9.778619 5.482538 12.794720 10.625030 7.128629ndata5 <- length(log2vsumasM3)** Primer histograma**
hist(log2vsumasM3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM3')
meanlog2vsumasM3 <- mean(log2vsumasM3); head(meanlog2vsumasM3)## [1] 6.317092StdDevlog2vsumasM3 <- sd(log2vsumasM3); head(StdDevlog2vsumasM3)## [1] 3.410882Normlog2vsumasM3 <- (log2vsumasM3-meanlog2vsumasM3)/StdDevlog2vsumasM3; head(Normlog2vsumasM3)## [1] 1.0148481 -0.2446740 1.8991065 1.2629984 0.2379259 -0.3441057tst<- Normlog2vsumasM3Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM3',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -2.967667e-17 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293print(CuantilesA)## LimInf LimSup
## 65 -0.8628952 0.8884901
## 70 -0.9319289 1.0207444
## 75 -1.0144701 1.1904247
## 80 -1.1171230 1.3814634
## 85 -1.2529569 1.6102025
## 90 -1.4542681 1.9460444
## 95 -1.8520407 2.4110529
## 99 -1.8520407 3.0342081Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
Grafica Cuantiles del \(65\%\) y \(80\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
Grafica Cuantiles del \(70\%\) y \(85\%\)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))